A Review on Computer Assisted Follicle Detection Techniques and Polycystic Ovarian Syndrome (PCOS) Diagnostic Systems

Abstract

Polycystic Ovarian Syndrome (PCOS) caused infertility in women if not diagnosed and treated early. Transvaginal ultrasound machine is a non-invasive method of imaging human ovary with the aim of revealing salient features necessary for PCOS diagnosis. Numbers of follicles and their sizes are the main features that characterize ovarian images. Hence, PCOS is diagnosed by counting the numbers of follicles and measuring their sizes manually. This process is laborious, prone to error and time consuming. This paper surveys various computer assisted techniques for the detection of follicles and PCOS diagnoses in the ultrasound images of the ovary. Performances of some of the previous works are identified and compared. Finally, future research directions to improve on some of the observed limitations are provided